DeepRoute.ai Unveils DeepRoute IO 2.0 with VLA Technology

DeepRoute.ai has launched DeepRoute IO 2.0, its latest smart driving platform powered by the Vision-Language-Action (VLA) model, announced in a press release. This platform aims to deliver smarter and safer driving experiences by integrating advanced reasoning and a comprehensive knowledge base.

DeepRoute IO 2.0 features a multi-chip, multi-sensor design that supports both LiDAR-equipped and pure vision configurations, allowing seamless integration across various vehicle models. The platform will debut on NVIDIA DRIVE AGX Thor, utilizing the Blackwell GPU architecture for enhanced reasoning capabilities.

The VLA model in DeepRoute IO 2.0 offers two core advantages: chain-of-thought reasoning and an extensive knowledge base. These features enable the system to analyze complex traffic scenarios and adapt to real-world conditions, enhancing the transparency and reliability of driving decisions. The platform also includes OCR capabilities for real-time traffic sign recognition and a voice control feature for natural language interaction.

DeepRoute.ai has secured five OEM partnerships for the deployment of DeepRoute IO 2.0, with the first production vehicles expected to hit the market later this year. The company continues to collaborate with global partners to accelerate the deployment of smart driving technology worldwide.